dragonfly.git
2 years agodma.8: dma(8) only uses two config files, auth.conf and dma.conf.
Sascha Wildner [Mon, 9 Jan 2012 09:15:42 +0000 (10:15 +0100)]
dma.8: dma(8) only uses two config files, auth.conf and dma.conf.

Submitted-by: Max Herrgard <herrgard@gmail.com>
2 years agomfiutil(8): Correct sysctl name.
Sascha Wildner [Mon, 9 Jan 2012 03:47:19 +0000 (04:47 +0100)]
mfiutil(8): Correct sysctl name.

2 years agoftp: NetBSD uses __dead, and we use __dead2.
Peter Avalos [Mon, 9 Jan 2012 01:46:12 +0000 (17:46 -0800)]
ftp:  NetBSD uses __dead, and we use __dead2.

2 years agoMerge branch 'master' of /home/www-data/gitweb/dragonfly
Peter Avalos [Mon, 9 Jan 2012 01:17:34 +0000 (17:17 -0800)]
Merge branch 'master' of /home/www-data/gitweb/dragonfly

2 years agoUpdate files for xz import.
Peter Avalos [Mon, 9 Jan 2012 01:14:04 +0000 (17:14 -0800)]
Update files for xz import.

This primarily installs some scripts and associated manual pages for
diff, grep, and less functionality for xz (xzdiff, xzgrep, xzless,
xzmore).

2 years agoisp.4: Use .Nx
Sascha Wildner [Sun, 8 Jan 2012 21:52:36 +0000 (22:52 +0100)]
isp.4: Use .Nx

2 years agoMerge branch 'vendor/XZ'
Peter Avalos [Sun, 8 Jan 2012 21:49:37 +0000 (13:49 -0800)]
Merge branch 'vendor/XZ'

2 years agoImport xz-5.0.3.
Peter Avalos [Sun, 8 Jan 2012 21:32:05 +0000 (13:32 -0800)]
Import xz-5.0.3.

    * xz --force now (de)compresses files that have setuid, setgid,
      or sticky bit set and files that have multiple hard links.
      The man page had it documented this way already, but the code
      had a bug.

    * LZMA2 decompressor now correctly accepts LZMA2 streams with no
      uncompressed data. Previously it considered them corrupt. The
      bug can affect applications that use raw LZMA2 streams. It is
      very unlikely to affect .xz files because no compressor creates
      .xz files with empty LZMA2 streams. (Empty .xz files are a
      different thing than empty LZMA2 streams.)

    * "xz --suffix=.foo filename.foo" now refuses to compress the
      file due to it already having the suffix .foo. It was already
      documented on the man page, but the code lacked the test.

    * liblzma fixes:

        - A memory leak was fixed.

        - lzma_stream_buffer_encode() no longer creates an empty .xz
          Block if encoding an empty buffer. Such an empty Block with
          LZMA2 data would trigger a bug.

        - Validate function arguments better in a few functions. Most
          importantly, specifying an unsupported integrity check to
          lzma_stream_buffer_encode() no longer creates a corrupt .xz
          file. Probably no application tries to do that, so this
          shouldn't be a big problem in practice.

        - Document that lzma_block_buffer_encode(),
          lzma_easy_buffer_encode(), lzma_stream_encoder(), and
          lzma_stream_buffer_encode() may return LZMA_UNSUPPORTED_CHECK.

        - The return values of the _memusage() functions are now
          documented better.

2 years agoMerge branch 'vendor/TNFTP'
Peter Avalos [Sun, 8 Jan 2012 20:43:07 +0000 (12:43 -0800)]
Merge branch 'vendor/TNFTP'

2 years agoftp: Import NetBSD's latest version.
Peter Avalos [Sun, 8 Jan 2012 20:36:27 +0000 (12:36 -0800)]
ftp:  Import NetBSD's latest version.

Move determination of socket buffer sizes from startup to the first
time a socket is used, as the previous logic assumed AF_INET sockets
were available (which they may not be in an IPv6-only system).

Use __dead

Fix warnings.

Obtained-from:  NetBSD

2 years agomfiutil(8): Use dehumanize_number() instead of rolling our own.
Sascha Wildner [Sun, 8 Jan 2012 19:03:57 +0000 (20:03 +0100)]
mfiutil(8): Use dehumanize_number() instead of rolling our own.

2 years agomfiutil(8): Remove some compatibility defines which we don't need.
Sascha Wildner [Sun, 8 Jan 2012 18:39:18 +0000 (19:39 +0100)]
mfiutil(8): Remove some compatibility defines which we don't need.

2 years agomfi(4): Fix volume creation with mfiutil(8).
Sascha Wildner [Sun, 8 Jan 2012 17:14:47 +0000 (18:14 +0100)]
mfi(4): Fix volume creation with mfiutil(8).

This change (taken from LSI in 175660922b2422bdda81a5967c5dadd176fe1c3c)
caused the mfi_config_data structure to change size which ultimately
ended up in the MFI_DCMD_CFG_ADD command failing. Stuff seems to work fine
without it.

Reported-by: Tomi Juvonen <tomi.juvonen@kenou.biz>
2 years agoahci: Black list more chips which do not support MSI properly
Sepherosa Ziehau [Sun, 8 Jan 2012 13:27:16 +0000 (21:27 +0800)]
ahci: Black list more chips which do not support MSI properly

According-to: Linux

2 years agopcidevs: regen
Sepherosa Ziehau [Sun, 8 Jan 2012 13:26:52 +0000 (21:26 +0800)]
pcidevs: regen

2 years agopcidevs: More device IDs for AHCI chips
Sepherosa Ziehau [Sun, 8 Jan 2012 13:26:27 +0000 (21:26 +0800)]
pcidevs: More device IDs for AHCI chips

2 years agovmstat: Always print interrupts cpuid under verbose mode
Sepherosa Ziehau [Sun, 8 Jan 2012 12:19:44 +0000 (20:19 +0800)]
vmstat: Always print interrupts cpuid under verbose mode

2 years agojme: Use MSI, if the device supports it
Sepherosa Ziehau [Sun, 8 Jan 2012 11:22:46 +0000 (19:22 +0800)]
jme: Use MSI, if the device supports it

2 years agodntpd: Don't assume that reconnect to server always succeeds
Sepherosa Ziehau [Sun, 8 Jan 2012 11:04:54 +0000 (19:04 +0800)]
dntpd: Don't assume that reconnect to server always succeeds

2 years agoUpdate the pciconf(8) database.
Sascha Wildner [Sun, 8 Jan 2012 03:16:11 +0000 (04:16 +0100)]
Update the pciconf(8) database.

Jan 6, 2012 snapshot from http://pciids.sourceforge.net/

2 years agomfi(4): Remove two NULL checks after kmalloc() with M_WAITOK.
Sascha Wildner [Sat, 7 Jan 2012 11:58:28 +0000 (12:58 +0100)]
mfi(4): Remove two NULL checks after kmalloc() with M_WAITOK.

2 years agomfi(4): Avoid some code duplication.
Sascha Wildner [Fri, 6 Jan 2012 18:06:48 +0000 (19:06 +0100)]
mfi(4): Avoid some code duplication.

2 years agomfi(4): Use callout_init_mp() for the watchdog callout.
Sascha Wildner [Fri, 6 Jan 2012 17:26:10 +0000 (18:26 +0100)]
mfi(4): Use callout_init_mp() for the watchdog callout.

mfi_timeout() is MP safe. When porting, I missed that FreeBSD passes
CALLOUT_MPSAFE here.

2 years agoioapic_abi/x86_64: Remove no longer applied warning
Sepherosa Ziehau [Fri, 6 Jan 2012 16:10:33 +0000 (00:10 +0800)]
ioapic_abi/x86_64: Remove no longer applied warning

While I'm here, strip blank lines

2 years agoahci: Create black list for devices which do not support MSI properly
Sepherosa Ziehau [Fri, 6 Jan 2012 15:49:47 +0000 (23:49 +0800)]
ahci: Create black list for devices which do not support MSI properly

2 years agopcidevs: regen
Sepherosa Ziehau [Fri, 6 Jan 2012 15:33:05 +0000 (23:33 +0800)]
pcidevs: regen

2 years agopcidevs: Add ATI SB7x0/SB8x0/SB9x0 AHCI device ID
Sepherosa Ziehau [Fri, 6 Jan 2012 15:32:36 +0000 (23:32 +0800)]
pcidevs: Add ATI SB7x0/SB8x0/SB9x0 AHCI device ID

2 years agoi386: Add MSI support for ICU MachIntrABI
Sepherosa Ziehau [Fri, 6 Jan 2012 15:20:55 +0000 (23:20 +0800)]
i386: Add MSI support for ICU MachIntrABI

This merges 759fc5332c2b97c3a48b5aefb94b6cab57de07a4

2 years agoi386: Add MSI support for IOAPIC MachIntrABI
Sepherosa Ziehau [Fri, 6 Jan 2012 14:54:59 +0000 (22:54 +0800)]
i386: Add MSI support for IOAPIC MachIntrABI

This merges:
4234567cdb0fd130f7b2aca6b09a0519348541eb
7b87350be06bbe28afbe743f8b189bcfce77c05a
and part of
2c3d7ac8b794c14f1a14b7cf04c0c144a1aed6dd

2 years agomsi/i386: Add setup/map function
Sepherosa Ziehau [Fri, 6 Jan 2012 14:02:18 +0000 (22:02 +0800)]
msi/i386: Add setup/map function

This merges c9852c4cdc5706115542e9c0f6228b2612e359d4

2 years agomsi/i386: Add interrupt vectors
Sepherosa Ziehau [Fri, 6 Jan 2012 13:52:45 +0000 (21:52 +0800)]
msi/i386: Add interrupt vectors

This is mainly based on apic/apic_vector.s

2 years agomfi(4) & mfiutil(8): Sync with FreeBSD.
Sascha Wildner [Fri, 6 Jan 2012 03:37:51 +0000 (04:37 +0100)]
mfi(4) & mfiutil(8): Sync with FreeBSD.

2 years agomfi(4): Use MSI if it is supported by the device.
Sascha Wildner [Wed, 4 Jan 2012 23:29:38 +0000 (00:29 +0100)]
mfi(4): Use MSI if it is supported by the device.

2 years agokernel: do not try to always build IPv4-specific code
Francois Tigeot [Thu, 5 Jan 2012 17:40:50 +0000 (18:40 +0100)]
kernel: do not try to always build IPv4-specific code

* it fails when no INET support is present

* and people may want to start running IPv6-only hosts

2 years agokernel: gre(4) should depend on 'options INET' if it is in the config.
Sascha Wildner [Thu, 5 Jan 2012 17:41:50 +0000 (18:41 +0100)]
kernel: gre(4) should depend on 'options INET' if it is in the config.

2 years agodfregress(8): Fix a typo in an error message.
Sascha Wildner [Thu, 5 Jan 2012 15:55:20 +0000 (16:55 +0100)]
dfregress(8): Fix a typo in an error message.

2 years agoisp(4): Use MSI if it is supported by the device.
Sascha Wildner [Wed, 4 Jan 2012 23:22:07 +0000 (00:22 +0100)]
isp(4): Use MSI if it is supported by the device.

2 years agoisp(4): Fix a typo in a message.
Sascha Wildner [Thu, 5 Jan 2012 15:05:35 +0000 (16:05 +0100)]
isp(4): Fix a typo in a message.

2 years agodma.8: Fix a few small issues.
Sascha Wildner [Thu, 5 Jan 2012 13:30:39 +0000 (14:30 +0100)]
dma.8: Fix a few small issues.

2 years agoUPDATING: Add some info about the dma(8) upgrade.
Sascha Wildner [Thu, 5 Jan 2012 13:30:10 +0000 (14:30 +0100)]
UPDATING: Add some info about the dma(8) upgrade.

2 years agopam: Need to list ruser for pam_group.
Peter Avalos [Thu, 5 Jan 2012 02:30:17 +0000 (18:30 -0800)]
pam:  Need to list ruser for pam_group.

2 years agoUpdate files for OpenSSL-1.0.0f import.
Peter Avalos [Thu, 5 Jan 2012 00:25:38 +0000 (16:25 -0800)]
Update files for OpenSSL-1.0.0f import.

2 years agoMerge branch 'vendor/OPENSSL'
Peter Avalos [Thu, 5 Jan 2012 00:20:06 +0000 (16:20 -0800)]
Merge branch 'vendor/OPENSSL'

2 years agoImport OpenSSL-1.0.0f.
Peter Avalos [Thu, 5 Jan 2012 00:06:12 +0000 (16:06 -0800)]
Import OpenSSL-1.0.0f.

      o Fix for DTLS plaintext recovery attack CVE-2011-4108
      o Clear block padding bytes of SSL 3.0 records CVE-2011-4576
      o Only allow one SGC handshake restart for SSL/TLS CVE-2011-4619
      o Check parameters are not NULL in GOST ENGINE CVE-2012-0027
      o Check for malformed RFC3779 data CVE-2011-4577

2 years agokernel: Use NULL for pointers.
Sascha Wildner [Tue, 3 Jan 2012 21:22:30 +0000 (22:22 +0100)]
kernel: Use NULL for pointers.

2 years agodma(8): Upgrade to v0.7.
Sascha Wildner [Tue, 3 Jan 2012 14:31:49 +0000 (15:31 +0100)]
dma(8): Upgrade to v0.7.

For the change log, see:

https://github.com/corecode/dma/blob/master/debian/changelog

Tested-by: luxh
Submitted-by: corecode
2 years agoRun VFS_START() for the root mount point.
Francois Tigeot [Thu, 25 Aug 2011 09:39:18 +0000 (11:39 +0200)]
Run VFS_START() for the root mount point.

The namecache has to be initialized properly for that. Use code
taken from start_init()

2 years agogold linker v2.22: Remove unnecessary assertions
John Marino [Tue, 3 Jan 2012 16:51:24 +0000 (17:51 +0100)]
gold linker v2.22: Remove unnecessary assertions

Fixes this error: internal error in relocate_tls
Reported as binutils bug 13359
<http://sourceware.org/bugzilla/show_bug.cgi?id=13359>
Applied to trunk 2011-10-31
Backported to 2.22 branch on 2011-12-19

2 years agogold linker v2.22: Fix constructor handling
John Marino [Tue, 3 Jan 2012 10:50:54 +0000 (11:50 +0100)]
gold linker v2.22: Fix constructor handling

Starting with gold 2.22, gold looks for contructors in DT_INIT_ARRAY by
default, but gcc defines them in a .ctors section.  Apparently the
presence of crtbegin and crtend with their .init sections are confusing
gold, but not gnu ld.

To combat this, a new option "--no-ctors-in-init-array" was added to
emulate gnu ld behavior of looking for constructors in the .ctors
section provided by gcc.  This commit makes this new option the default
behavior on DragonFly's gold.

Without this modification, a gold-built world is unbootable!

2 years agokernel - Fix a case in if_re which could lockup the system
Matthew Dillon [Tue, 3 Jan 2012 17:18:28 +0000 (09:18 -0800)]
kernel - Fix a case in if_re which could lockup the system

* if_re's re_start function was not always properly setting IFF_OACTIVE
  in situations where packets would be left on the if_snd queue, causing
  if_start_dispatch() to get into an endless message loop.

* Add debugging to device_printf() when the case occurs.

2 years agoispfw(4): Add MODULE_VERSION()s.
Sascha Wildner [Tue, 3 Jan 2012 14:36:35 +0000 (15:36 +0100)]
ispfw(4): Add MODULE_VERSION()s.

2 years agoispfw(4): Correct case in module name.
Sascha Wildner [Tue, 3 Jan 2012 13:52:47 +0000 (14:52 +0100)]
ispfw(4): Correct case in module name.

2 years agokernel/vfs: enable accounting for ro mounts
François Tigeot [Tue, 3 Jan 2012 13:10:16 +0000 (14:10 +0100)]
kernel/vfs: enable accounting for ro mounts

* Some filesystems may be first mounted read-only and then upgraded
  to read-write status

* The VFS layer has no provision for signaling VFSes of status change,
  mount-point specific initialization code can only be run for the
  initial mount

* VFS accounting was thus never enabled on a ro->rw mount point transition

* Bite the bullet and run vq_init() unconditionally for both read-only
  and read-write mounts, the cost is only a few variables set to NULL
  or zero anyway.

2 years agoisp(4): Sync with FreeBSD.
Sascha Wildner [Tue, 3 Jan 2012 12:29:09 +0000 (13:29 +0100)]
isp(4): Sync with FreeBSD.

Adds support for some newer cards:

* Qlogic ISP 2322 PCI FC-AL Adapter

* Qlogic ISP 2422 PCI FC-AL Adapter

* Qlogic ISP 2432 PCI FC-AL Adapter

* Qlogic ISP 2532 PCI FC-AL Adapter

* Qlogic ISP 5432 PCI FC-AL Adapter

* Qlogic ISP 6312 PCI FC-AL Adapter

* Qlogic ISP 6322 PCI FC-AL Adapter

Also isp(4) can now be loaded as a module. The firmwares have been
separated into per-firmware modules, too.

It was tested with an old DEC KZPBA-CX (1040 based) SCSI adapter.

2 years agokernel/mchain: Fix a NULL check.
Sascha Wildner [Tue, 3 Jan 2012 09:22:00 +0000 (10:22 +0100)]
kernel/mchain: Fix a NULL check.

2 years agokernel - Fix swapcache inefficiency related to the PQ_INACTIVE queue
Matthew Dillon [Tue, 3 Jan 2012 05:25:47 +0000 (21:25 -0800)]
kernel - Fix swapcache inefficiency related to the PQ_INACTIVE queue

* The swapcache was still coded with the assumption of just one
  PQ_INACTIVE queue.  Even though it scans all 256 queues the
  end-of-queue code could trigger early and prevent later queue
  indices from being scanned, causing a large percentage of eligible
  VM pages to not get written to the swapcache.

  Now all queues must signal that they have reached the end before the
  code sets the end-of-queue condition and loops up to the 1/10 second
  sleep.

* The full count/scount was being applied to each queue individually instead
  of to all queues uniformly, allowing more pages to be laundered than
  desired.  Divide out and add a bit of slop.

2 years agokernel - Use atomic ops to update hammer_count_refedbufs
Matthew Dillon [Mon, 2 Jan 2012 19:27:56 +0000 (11:27 -0800)]
kernel - Use atomic ops to update hammer_count_refedbufs

* This global is used only for statistical reporting via sysctl and has
  no operational effect.

* Use atomic ops to avoid tracking errors from multiple hammer mounts.

2 years agohammer - Pass -S option to remote mirror-read
Matthew Dillon [Mon, 2 Jan 2012 18:10:52 +0000 (10:10 -0800)]
hammer - Pass -S option to remote mirror-read

* When doing a mirror or mirror-stream via pull from the remote pass the
  -S option to the remote end, otherwise the remote will just use the
  default split size.

Reported-by: ftigeot
2 years agoktr.9: Adjust for the recent ktr work by aggelos.
Sascha Wildner [Mon, 2 Jan 2012 16:48:23 +0000 (17:48 +0100)]
ktr.9: Adjust for the recent ktr work by aggelos.

2 years agotest/cocci/NELEM.cocci: Oops, wrong comment character.
Sascha Wildner [Mon, 2 Jan 2012 02:15:26 +0000 (03:15 +0100)]
test/cocci/NELEM.cocci: Oops, wrong comment character.

2 years agoExpand the comments in the coccinelle patches a bit more.
Sascha Wildner [Mon, 2 Jan 2012 02:10:46 +0000 (03:10 +0100)]
Expand the comments in the coccinelle patches a bit more.

2 years agoAdd a coccinelle patch to find missing lock release before return.
Sascha Wildner [Mon, 2 Jan 2012 02:03:24 +0000 (03:03 +0100)]
Add a coccinelle patch to find missing lock release before return.

2 years agokernel/i4b: Instead of returning, goto fail (for crit_exit()).
Sascha Wildner [Mon, 2 Jan 2012 01:28:15 +0000 (02:28 +0100)]
kernel/i4b: Instead of returning, goto fail (for crit_exit()).

2 years agodpt(4): Add a missing crit_exit().
Sascha Wildner [Mon, 2 Jan 2012 01:27:17 +0000 (02:27 +0100)]
dpt(4): Add a missing crit_exit().

2 years agokernel/disk drivers: Add some missing crit_exit()s.
Sascha Wildner [Mon, 2 Jan 2012 01:26:29 +0000 (02:26 +0100)]
kernel/disk drivers: Add some missing crit_exit()s.

2 years agoed(4): Replace a forgotten crit_exit() with lwkt_serialize_exit().
Sascha Wildner [Mon, 2 Jan 2012 00:06:34 +0000 (01:06 +0100)]
ed(4): Replace a forgotten crit_exit() with lwkt_serialize_exit().

Only cosmetical as this code is not enabled.

2 years agovkernel - Unbreak 32-bit vkernel builds by fixing broken assertion
Matthew Dillon [Sun, 1 Jan 2012 23:08:37 +0000 (15:08 -0800)]
vkernel - Unbreak 32-bit vkernel builds by fixing broken assertion

* [ASSERT_]MP_LOCK_HELD() was not properly handling the new token
  t_count format.

* Rename and document the functions a bit better.  Currently our
  more expansive token test for (shared or exclusive) is not
  a perfect test because we don't want to eat cpu time iterating
  the thread's token array to check for shared tokens.  So we just
  check for a shared count.

  Works well for assertions, wouldn't work very well for conditional
  acquisition (of a shared token).

* Exclusive token tests are perfect.

Reported-by: tuxillo
2 years agokernel - Fix ENOBUFS related panic in tcp_output
Matthew Dillon [Sun, 1 Jan 2012 20:14:40 +0000 (12:14 -0800)]
kernel - Fix ENOBUFS related panic in tcp_output

* If the tcp_output code gets all the way to the end but then has to
  'goto again', it fails to NULL out the (th) pointer.  If the mbuf fails
  on the retry (m == NULL), the non-NULL (th) winds up causing the code to
  believe that the mbuf is good and results in a null-pointer dereference.

2 years agohammer(8): fix an obsolete comment
François Tigeot [Sun, 1 Jan 2012 15:01:30 +0000 (16:01 +0100)]
hammer(8): fix an obsolete comment

2 years agokgdb: Fix double-free bug
John Marino [Sun, 1 Jan 2012 12:00:29 +0000 (13:00 +0100)]
kgdb: Fix double-free bug

It was a mistake to reuse target sections for each inferior.  Leaving the
default section for each program space eliminates the redundant xfrees.
The exec.c function "resize_table_sections" code was painful to look at with
unused variables and unnecessary memory reallocations, so it was altered
at the same time.

2 years agokernel/tws: Add a missing lockmgr(...,LK_RELEASE).
Sascha Wildner [Sun, 1 Jan 2012 01:48:53 +0000 (02:48 +0100)]
kernel/tws: Add a missing lockmgr(...,LK_RELEASE).

2 years agokernel/gpio: Add a missing lockmgr(...,LK_RELEASE).
Sascha Wildner [Sun, 1 Jan 2012 01:47:24 +0000 (02:47 +0100)]
kernel/gpio: Add a missing lockmgr(...,LK_RELEASE).

2 years agokernel/vfs_quota: Add a missing spin_unlock().
Sascha Wildner [Sun, 1 Jan 2012 01:31:11 +0000 (02:31 +0100)]
kernel/vfs_quota: Add a missing spin_unlock().

2 years agokernel/netgraph: Add a missing lwkt_reltoken().
Sascha Wildner [Sun, 1 Jan 2012 01:02:26 +0000 (02:02 +0100)]
kernel/netgraph: Add a missing lwkt_reltoken().

2 years agokernel/cam: Add two missing lockmgr(...,LK_RELEASE)'s.
Sascha Wildner [Sun, 1 Jan 2012 00:35:16 +0000 (01:35 +0100)]
kernel/cam: Add two missing lockmgr(...,LK_RELEASE)'s.

2 years agoBump copyright. Happy new year!!
Sascha Wildner [Sat, 31 Dec 2011 23:51:55 +0000 (00:51 +0100)]
Bump copyright. Happy new year!!

2 years agoMerge branch 'vendor/FILE'
Peter Avalos [Sat, 31 Dec 2011 20:44:00 +0000 (12:44 -0800)]
Merge branch 'vendor/FILE'

2 years agoImport file-5.10.
Peter Avalos [Sat, 31 Dec 2011 20:32:20 +0000 (12:32 -0800)]
Import file-5.10.

Changes are documented in ChangeLog.

2 years agoReduce differences between our VKERNEL and VKERNEL64 configurations.
Sascha Wildner [Sat, 31 Dec 2011 11:36:16 +0000 (12:36 +0100)]
Reduce differences between our VKERNEL and VKERNEL64 configurations.

* Removes KTR from VKERNEL.

* Removes VM_PAGE_DEBUG from VKERNEL64.

* Some whitespace adjustments.

2 years agoVKERNEL64: Adjust for the ktr changes (to fix build).
Sascha Wildner [Sat, 31 Dec 2011 11:17:55 +0000 (12:17 +0100)]
VKERNEL64: Adjust for the ktr changes (to fix build).

2 years agolibdl -- Stub libdl.
Venkatesh Srinivas [Sat, 31 Dec 2011 10:21:05 +0000 (05:21 -0500)]
libdl -- Stub libdl.

DragonFly BSD (and FreeBSD) have the dlopen/dlfcn/... routines located in
libc. However many applications and their configure scripts look for these
routines in libdl. This adds a stub libdl with no subroutines, so that the
configure scripts pass. The dl* routines are automatically available via
libc.

2 years agokernel -- nfsserver: Allow for some reordering of RPCs before decaying seqcount.
Venkatesh Srinivas [Sat, 31 Dec 2011 05:27:01 +0000 (00:27 -0500)]
kernel -- nfsserver: Allow for some reordering of RPCs before decaying seqcount.

Before decaying sequential heuristic, allow a limited amount of RPC reordering.
After that, decay by half for each out-of-order RPC rather than immediately
dropping to 1.

From FreeBSD r228185.

2 years agokernel -- KTR: Fix vkernel32 build.
Venkatesh Srinivas [Sat, 31 Dec 2011 05:21:51 +0000 (00:21 -0500)]
kernel -- KTR: Fix vkernel32 build.

KTR_INFO needed to be updated for the typesafe KTR patchset.

2 years agokernel: Fix two NULL dereference issues.
Sascha Wildner [Sat, 31 Dec 2011 03:38:52 +0000 (04:38 +0100)]
kernel: Fix two NULL dereference issues.

2 years agokernel/acpi: Add a missing file to the 'files' files (fixes LINT* build).
Sascha Wildner [Sat, 31 Dec 2011 03:37:30 +0000 (04:37 +0100)]
kernel/acpi: Add a missing file to the 'files' files (fixes LINT* build).

2 years agovkernel/intr: Pin vkernel "hardware interrupts" to CPU0
Sepherosa Ziehau [Sat, 31 Dec 2011 03:29:32 +0000 (11:29 +0800)]
vkernel/intr: Pin vkernel "hardware interrupts" to CPU0

Due to the nondeterminsm of CPU to which SIGIO will be delivered,
we can't depend on "hardware interrupts" of vkernel always happen
on the registered CPU.  We now pin the vkernel "hardware interrupts"
to CPU0, so the interrupt scheduling code could always know the
correct ithread to be scheduled.

Reported-by: tuxillo vsrinivas
2 years agoAdjust ktrdump(8) for the ktr(4) changes
Aggelos Economopoulos [Fri, 30 Dec 2011 13:44:21 +0000 (05:44 -0800)]
Adjust ktrdump(8) for the ktr(4) changes

- On i386, we need to extend logged integers to be at least
  int-sized like vprintf expects.
- On x86_64, we unpack the logged arguments into a va_list according
  to the x86_64 ABI rules and then pass it to kvprintf.

2 years agoModify ktr(4) to be typesafe
Aggelos Economopoulos [Fri, 30 Dec 2011 13:38:34 +0000 (05:38 -0800)]
Modify ktr(4) to be typesafe

- Use a struct copy to place the logged values directly into the
  destination buffer without forcing a memcpy.
- This also fixes event logging on x86_64, as we no longer rely
  on the i386 calling convention for functions with a variable
  number of arguments.
- Verify that the arguments match the specified format.
- Update callsites.

Note that this requires userspace changes! ktrdump(8) needs to be
adjusted even on i386 (%c and %hd are no longer promoted as per the
C rules). We need very special handling on x86_64 too.

2 years agoIntroduce sys/cpputil.h for our cpp metaprogramming needs
Aggelos Economopoulos [Fri, 30 Dec 2011 17:00:44 +0000 (09:00 -0800)]
Introduce sys/cpputil.h for our cpp metaprogramming needs

2 years agokernel/drm: Move a NULL check to the proper place.
Sascha Wildner [Fri, 30 Dec 2011 16:39:11 +0000 (17:39 +0100)]
kernel/drm: Move a NULL check to the proper place.

2 years agokernel/altq: Move a dereference below the NULL check.
Sascha Wildner [Fri, 30 Dec 2011 16:13:32 +0000 (17:13 +0100)]
kernel/altq: Move a dereference below the NULL check.

2 years agoacpi/cpu: Missing one 'O' in the macro name
Sepherosa Ziehau [Thu, 29 Dec 2011 11:01:27 +0000 (19:01 +0800)]
acpi/cpu: Missing one 'O' in the macro name

2 years agoacpi/cpu: Report that Turbo Boost is enabled on Intel CPUs
Sepherosa Ziehau [Thu, 29 Dec 2011 10:36:41 +0000 (18:36 +0800)]
acpi/cpu: Report that Turbo Boost is enabled on Intel CPUs

If Turbo Boost is enabled, the current CPU frequencies reported by
hw.acpi.cpu are not correct, sigh.

2 years agoacpi/pstate: Print coordination
Sepherosa Ziehau [Thu, 29 Dec 2011 09:26:44 +0000 (17:26 +0800)]
acpi/pstate: Print coordination

2 years agoacpi/pstate: Fix the long standing P-State detection problem on Intel CPUs
Sepherosa Ziehau [Thu, 29 Dec 2011 08:33:11 +0000 (16:33 +0800)]
acpi/pstate: Fix the long standing P-State detection problem on Intel CPUs

- Rename the ACPI_CAP to ACPI_PDC according to Intel's document.
  While I'm here, update the document revision.
- Remove unapplied comment about _OSC and _PDC's revision; Intel's
  document states clearly that it should be 1
- Fix the 'Count' argument for _OSC evaluation; the intergers in
  'Capabilities buffer' argument is 2
- Fix the buffer length of _OSC's 'Capabilities buffer'; we only
  pass 2 intergers.
- Perfer _OSC evaluation, fall back to _PDC evaluation only if _OSC
  evaluation fails.
- Add MD cpu features, so MD code could deliver proper settings.
  For AMD CPUs, the old configuration just works (AMD actually has
  no documents about _PDC and _OSC).
  For Intel CPUs w/ EST, it looks like P-State's _PCT will appear
  only when bit0 (P-State MSR), bit5 (P-State software coordination)
  and bit11 (P-State hardware coordination) are turned on.

2 years agox86_64: Add MSI support for ICU MachIntrABI
Sepherosa Ziehau [Thu, 29 Dec 2011 05:56:31 +0000 (13:56 +0800)]
x86_64: Add MSI support for ICU MachIntrABI

2 years agointr: Remove no longer correct ithread_cpuid; use rman_get_cpuid instead
Sepherosa Ziehau [Thu, 29 Dec 2011 03:04:19 +0000 (11:04 +0800)]
intr: Remove no longer correct ithread_cpuid; use rman_get_cpuid instead

2 years agolibmagic: Update Makefile.
Peter Avalos [Wed, 28 Dec 2011 22:40:49 +0000 (14:40 -0800)]
libmagic:  Update Makefile.

-Use SHAREDIR from bsd.own.mk instead of hardcoding it.
-Get rid of -I nonsense that was required when we used to patch the
vendor's .h, but we need to turn off warnings when building mkmagic.nx.

Inspired-by: http://bugs.dragonflybsd.org/issues/2206
2 years agobce: Add workaround for the lost of MSI on 5706/5708
Sepherosa Ziehau [Wed, 28 Dec 2011 14:44:08 +0000 (22:44 +0800)]
bce: Add workaround for the lost of MSI on 5706/5708

While I'm here use ramn_get_cpuid to extract interrupt's target CPU

Obtained-from: bnx2

2 years agoem: rman_get_cpuid should be used to get the interrupt's target CPU
Sepherosa Ziehau [Wed, 28 Dec 2011 13:53:41 +0000 (21:53 +0800)]
em: rman_get_cpuid should be used to get the interrupt's target CPU